复杂施工条件下无人碾压机群协同全覆盖路径规划研究  

Research on cooperative complete coverage path planning for unmanned roller group under complex construction conditions

作　　者：时梦楠 崔博[1] 王佳俊[1] 关涛[1] 佟大威[1] 任炳昱[1] SHI Mengnan;CUI Bo;WANG Jiajun;GUAN Tao;TONG Dawei;REN Bingyu(Tianjin University,State Key Laboratory of Civil Engineering Simulation and Safety,Tianjin 300350,China)

机构地区：[1]天津大学水利工程仿真与安全国家重点实验室,天津300350

出　　处：《水利学报》2020年第12期1544-1557,共14页Journal of Hydraulic Engineering

基　　金：国家自然科学基金雅砻江联合基金(U1965207);中国工程院咨询项目(2019-XY-24);国家自然科学基金项目(51779169)。

摘　　要：无人碾压是大坝碾压施工领域研究的热点。大坝施工具有坝面面积大、料区多、障碍物多等复杂特点,这给无人碾压机群协同作业带来挑战。研究无人碾压机群高效协同全覆盖路径规划对于保证无人碾压机在大坝复杂施工条件下的碾压质量和进度具有重要意义。本文提出一种无人碾压机群协同全覆盖路径规划的新方法,包括作业面分解、子作业面规划、子作业面连通和机群任务分配等步骤。首先,针对存在障碍物的不规则边界作业面分解问题,采用Boustrophedon算法将其拆分成较少的无障碍物子作业面;其次,提出满足搭接法和错距法碾压施工原理的碾压运动模型(CMM,Compaction Motion Model)来进行子作业面路径规划;再者,将各子作业面连通施工顺序优化问题建模为TSP模型(TSP,Traveling Salesman Problem),采用混沌理论改进蜻蜓算法的搜索机制,构建混沌蜻蜓算法,以求解较优的子作业面间转场路径;最后,建立无人碾压机群作业时间成本函数,并以最小化时间成本为准则将全覆盖路径分配至各无人碾压机,实现无人碾压机群的高效协同全覆盖路径规划。以心墙堆石坝为例,通过Matlab仿真对比和工程应用,结果表明本文所提出的方法能够实现复杂施工条件下的无人碾压机群的协同全覆盖最优作业路径规划。同时,在碾压质量方面能够满足各料区碾压达标率均高于95%,在碾压效率方面无人碾压机设备利用率均高于97%。本研究为大坝碾压机群无人驾驶协同全覆盖路径规划提供了一种新思路。Unmanned roller is a hot topic in the field of dam construction.Dam construction has the complex characteristics of large dam face area,many material areas,and many obstacles,which brings challenges to the cooperative operation of unmanned roller compactor groups.Studying the efficient and cooperative complete coverage path planning of unmanned roller groups is of great significance to ensure the rolling quality and progress of unmanned roller under complex construction conditions.This paper proposes a new method for unmanned roller group collaborative complete coverage path planning,including steps such as block decomposition,sub-block planning,sub-block connectivity and task assignment.First,for the problem of irregular boundary working surface decomposition with obstacles,the Boustrophedon algorithm is used to split it into fewer unobstructed sub-block;Second,a CMM model(CMM,Compaction Motion Model)that meets the rolling construction standard of overlapping method and offset method is proposed to plan the path of the sub-block;Moreover,the sub-block connect problem is modeled as a TSP model(TSP,Traveling Salesman Problem).The chaos theory is used to improve the search mechanism of the dragonfly algorithm,and the chaotic dragonfly algorithm is constructed to solve the optimal transition path between sub-blocks;finally,establish the operation time cost function of the unmanned roller groups,and allocate the complete coverage path to each unmanned roller based on the principle of minimizing time cost To realize efficient and coordinated complete coverage path planning for unmanned roller groups.Taking the core rockfill dam as an example,through simulation and engineering application,the results show that the method proposed in this paper can not only avoid obstacles,but also achieve collaborative complete coverage optimal operation path planning for unmanned roller groups.At the same time,in terms of compaction quality,it can meet the compaction rate of each material area is higher than 95%,and in terms of rollin

关 键 词：复杂施工条件 无人碾压机群 协同全覆盖路径规划 Boustrophedon算法 碾压运动模型(CMM) 混沌蜻蜓算法 任务分配 

分 类 号：TV53[水利工程—水利水电工程]